Getting around

I'm currently attending a First Nations Technology summit because of some course work completed last fall. It is unrelated to my thesis, but instead of listening to any of the presentations this morning, I spent the time thinking about what my next steps need to be and trying to work out some of the bugs.

I've been trying to reconcile my working methods with time constraints, and I've realized that while I want to be efficient, time spent working on things that I do not present is not neccessarily a waste of time. While I don't have a lot to show for the work I have done since my last committee meeting, I do think that I'm now at a place where I can go into production mode and start to get drawings, diagrams and models that describe what I've been working on this semester.

So, this is what I currently intend to produce for next week:

1. 3-D models that combine each of the culls, in a matrix
2. Section cuts from the centre of each of the cuts
3. A diagram that compares this section cut to the original Architectural Graphic Standards and shows how the 3-D model can be altered to match the AGS section
4. Diagrams that show the four sides of each of the 3-D models of combined culls and compares them for combinations

This is a ton of work, but I think it's going to be useful. I think that removing the envelope information from the Ambiguous Sectional Conditions (once called MA's) was a good use of time because I can reintroduce it after I assemble groups of 3-D models. Hopefully I won't have to cull again: I don't think that there will be many AGS' that can be applied to the 3-D models, so hopefully that won't produce too many options, and I'll focus on one hub that combines different sides.

When I have done this I'm going to need to produce a full lineage diagram that tracks exactly how I got each hub: the 'parents' of each level. I think I'll do this last!

I need to get the 3-D models done ASAP, so I'm going to find a quiet corner at this conference where I can get down and spend some time working in Maya.

Still here...

I'm still at school. I'm doing a presentation downtown tomorrow and will miss more than half the day of work, so I need to get something finished tonight.

I've been thinking about the ways I've been culling the ambiguous sectional moments, and decided that it was worth another look. I want to produce another matrix that deals with environment (like envelope, shelter from rain, shelter from sun and accessibility), so I've removed all of the 'envelope' category from the previous group. This means that I don't have as many options, but at this point that is a good thing.

I've re-culled the remaining sections by picking a representative from each kind of combination. If I take them into Maya and loft them, I can then look at how they can address environmental conditions.

The architect of my own destruction?

I spent a little while last night thinking about what my next step should be for my midterm and I realized that I haven't fully made the transition out of the ambiguous sectional conditions into three-dimensions. I'm still looking for how and why the sections should be assembled.

Since I don't want to apply programs yet and it's also too early for environmental conditions, I thought I would go ahead and do another matrix. I went ahead and culled them, getting a total of 17 conditions, so I'm going to go ahead and cross each of them together and look at what that produces. I can then maybe look at combining them in the x and y direction, adding weather protection, and think about normalizing the slope. There are certain ramifications of altering the product of two conditions to get a 1:12 slope for accessibility, or a steeper slope for seating and I don't want to make those decisions lightly.

On a side note, it's been strange working this week because since it is reading week there are very few people in the studio. Even the keeners seem to be working from home, so it's chilly and a little lonely here.

Environmental stats

So, following the Flame's comments about environment and my project, I do think that it's important to include some environmental factors in this scale jump. I also want to include program stuff in scale. I might need to have two scale shifts: one for environment and one for program. See, I don't think that I can assemble multiple ambiguous spatial conditions for program yet because I don't want to limit the program by specifying stuff. I realize I'll have to do it eventually, but I think that if I leave it later I'll be able to retain more possible programs. Another reason is that it is too much of a jump between the sectional conditions and program. So introducing environmental factors could be a way to bridge that gap.

Environmental factors (all stats from Environment Canada):

Sun.
In winter, we want sun, in summer we don't. The most popular way to get both is to make an overhang that is deep enough to exclude summer sun but not so deep it keeps out low angle winter sun. In Vancouver, the sun is at around 67.5 degrees on June 21. To keep it off the floor of a 8' room, you need an overhang of 3'-4" at 8 feet. The higher you get, the larger the overhang needs to be, at a rate of about 5" per foot. This keeps out sun only June 21st, the summer solstice. Vancouver is hottest about August, so its a good idea to keep the sun out until then.

Rain.
We have more rain in the winter than in the summer, and frankly it's much less pleasant to be wet in the winter than it is in the summer. And while you might expect the overhang to keep off the rain, most of the time in Vancouver rain comes with wind. Which means it rains at an angle so roof cover doesn't always help.

Environment Canada says that in winter, the wind usually comes from the east and blows around 7-7.4 km/hr from the east (or 2 m/s). Small drops of rain fall around 6.5 m/s (or 23.4 km/hr) and large ones about 9 m/s (or 32.4 km/hr). So for our mythical 8' room, we would need an overhang of 1'-10" for small raindrops and 2'-5" for large ones, or between 2.5" and 3.6" for each foot of room on the east side of the building.

The problem is that while that is the average, the maximum hourly is 32 km/hr (which is around 9 m/s) from the north. This means that our 8' room will need an overhang of 8' on the north to keep off all potential rain.

Wind
It sucks when it's cold and windy, but sometimes when it is sunny if you can find a space out of the wind it can be lovely. The fortunate thing about wind is that if you are protected on a couple of sides, you are usually fine. These two sides would likely be those of prevailing breezes: the north and east. Micropatterns exist around buildings, but it doesn't take a lot of shelter to get you out of the wind.

The problem with environmental stuff is it is so directional. I don't want to decide what direction a seat will be facing if I don't know where it is going to go. Stupid catch 22.

ETA: some other stuff: any grassy area need to be extremely well drained. If it gets sun during the day it will be extremely popular with students. IF there is a roof on whatever thing gets produced, it should be accessible.

Using and abusing the chosen few

I've culled my ambiguous spatial conditions: I'm down to 15 from 115, so I think that's useful. I started by getting rid of the conditions that only matched one surface (about a quarter of them) and then I sorted the rest into types and picked the most interesting of each type -- there are a couple with two, but in most cases each type has a single example. I'm going to leave the rest of them for now, with the idea that I chose what I found to be interesting, but there is always the possibility of going back to the well, so to speak.

I spent some time today putting together a use diagram for the bus loop area. This is the diagram from FOA's Yokohama Port Terminal:

This is mine:
Not as interesting, not as dynamic. I'm also not entirely clear how this informs my scale move. I know there are lots of possibilities for things to add: graduation, meeting friends and studying, suntanning and people watching, political meetings. I have a list of all the possible uses, so maybe what I need to do is imagine how many people will want to do each of them at each part of the day. This one starts with the bus loop stuff, which I think is important, but my other use ideas don't really fit with it.

I also need to start looking at what additional factors should be considered in putting together the spatial hubs. Should I make examples that deal with rain/wind/sunlight? There must be a way to integrate shelter without just adding something.

I've been reading some of FOA's stuff today, and my favorite part is the way they use grass to cover slopes over 12%. I think this is awesome, and a great idea to indicate other kinds of uses. I think material fits in here somewhere as well. It's almost environmental: it's more pleasant to sit on a concrete bench in the sun, but a wood bench when it is damp or cold, even if neither of them are wet. So not only should I be arranging ambiguous spatial conditions to protect from/take advantage of sun, rain, wind, cold and hot, but I should maybe also look at what materials can do this.

So maybe I need to start making rules about these things. I have some climate information for Vancouver that shows the percentage of possible daylight hours for months of the year. If there's a 21% chance of sunshine in January, then there is a 79% chance of rain, so almost 80% of the area should be covered or otherwise protected. Maybe I should look at my chosen few to see what parts can also serve as shelter. Yeah, I think I'll do that and get back to you.

The first one's a doozy...

So, next steps:

Look at sorted ambiguous spatial conditions. Cull: determine which should be kept and which need to go. 15-20 is probably the max to start with, although others could be gradually reintroduced as required.

Produce Maya combinations for all ambiguous spatial conditions: remember, project is most creative when not taken too seriously. Have fun!

Move up scale: look at mechanism for combining conditions. Begin to consider general site: rain, sun, ground plane, envelope, doors, etc. Possibility of a matrix to consider options?

Produced 24 hr. use diagram for site. Look at FOA for precedent.

Assemble midterm presentation. Identify key points. Remember to begin meeting with the intent and idea of project. Clarity is key.

Project intent

In my thesis I would like to investigate the effects of small scale articulations on underprogrammed and empty space. Warehouses and open plazas often do not meet the needs of users because they do not incorporate a human scale. By beginning with human dimensions, ambiguous spatial conditions can begin to indicate use without dictating program. In addition they provide possibilities for users to occupy spaces as they see fit.

In order to test the robustness of ambiguous spatial conditions they will be applied to the programmatically diverse transit plaza intended for the UBC Student Union Building area. The open exterior plaza and food court are both possible sites for design, as each is required to support many different programs simultaneously.

Note: This project and my vocabulary no longer include the terms 'kit of parts', 'tool', 'deploy' or 'useful assembly'.

Post committee meeting update

I've spent a significant part of this week working on a paper for another course and trying to make up for the big work push from last weekend. But I'm back on thesis duty full time until the midterm review, which happens Feb 28th.

I was extremely happy with the last meeting. On Monday night I was finally able to spatialize the MA's (which will hereafter be known as ambiguous spatial conditions) by using Maya to loft between three conditions, producing this:

The consensus among my committee members was this is a productive direction and it's time for me to change scale and move up to what I think of as car size: find rules for assembling these guys.

I was shocked to find out that Thesis Advisor M, Big D and the Flame are unable to read my mind. They pointed out to me that the final intent of the project is still unclear to them and that I should really state that at the beginning of my presentation. I think one of the biggest things for the midterm is not so much that I need to move forward with my project as I need to find a way to clearly show what I've been working on, particularly with my DS, so that everyone watching the presentation is clear about where I am going. I'm totally going to rope Team into critiquing what I show so everything is clear. This is the first time I'll be presenting to people who haven't read my DS so it's a great experience, but I never know what's important about my own project.

The idea of a 24 hour diagram of use for the site came up again, so I'll try to produce one for the next meeting, and Thesis Advisor M suggested I be more specific about what I'm looking to critique in my site. I really enjoyed the conversation about how my project subverts all these givens in Vancouver architecture: it isn't anything I was thinking about at all, but I can see how it would apply. It's strange to have people ascribing motives to you that you haven't considered, but I'd like to play with some of the ideas and see what comes up.

Next post: intent!

Proxemics III

As much as I have enjoyed reading about proxemics, the time I've spent looking into it is wasted if I can't identify and clearly explain why I think it's related to my thesis. Why should I care how close people like to be to each other?

Well, what have I done since my last committee meeting?

What I have done since my last committee meeting

By Archigeek

The last time we saw Archigeek's thesis committee, they commented that her site research seemed premature, and that she think about her definition of ambiguity. Her notes suggest she create a 24hr flow diagram of the site and look at scales of occupations, linear propogation and exponential growth.

Well, since my last committee meeting, I have been thinking about definitions of ambiguity. I realized that the microarchitectural articulations (MA's) were not covering enough potential uses, and the way they were derived was unsystematic. I decided that if I was going to seriously consider MA's as a process, I should investigate all the different surface possibilities inherent in the 16 MA's I had created.

Part of the reason I did this was that the MA's were still program driven. Because they came from Architectural Graphic Standards (AGS), they still appeared to be very linked to the uses keyed to different positions of the body. I was also very aware that when I had been putting the MA's together I had been thinking of different programs that they could be applied to.

I then created a matrix that could explore some other possible MA's. I took each of the initial drawings from AGS: ramp, stair, sitting reach, standing reach, reclining angles, facing table, high work counter, and stage and added envelope as it had appeared numerous times in the MA's. I 'bred' each of these 9 sections together: this significantly increased the MA's. As some of the original MA's had three antecedents, I did this a second time to derive second generation MA's.

I now have a full tool kit of sectional moments. Because I bred together all possibilities, some uses work better than others and in some cases the original uses are impossible (or as Team pointed out, extremely dangerous). This is not a problem: in fact, it may even be desireable because it removes program from the sections and reduces them to an investigation of coincidences in surface variation keyed to the human body.

The matrix opens up my definition of ambiguity by allowing each moment to do more than the 2 or 3 AGS that spawned it. They are no longer designed to be ambiguous because the user needs to choose from 2 or 3 possible uses, but because they appear so strange there is no clear use and so the user need to choose from an infinite amount of habitations.

While the matrix has been tremendously useful, the MA's are still stuck as sectional moments. This is where the study of proxemics starts to come into play. While all the AGS have mimum widths (for example, a seat must be 16" or wider to be used as a seat) most do not have maximums. Those that do are linked to material -- they need to be narrow enough for users to slide their legs into the leg space. This is the case for the facing table and the desk.

So what criteria should be used for giving the MA's depth? I don't think that they can be given depth without beginning to program them. Instead, I looked at human spatial bubbles to find what kind of options I have for width. I'm still looking small, so I decided that a group of two people would be appropriate (something that I should also note is that in a study of 7,405 informal groups and 1,458 people working, 71% of all groups were two people). I took three MA's as examples and applied human bubble distances to them to see what happens. It becomes clear that planning for two people who want to be in each others' intimate or personal space is fine, when they wish to remain at a social or public space, density is too low to design for a public space.

This is when I began to look at the strategies people use to withdraw when people are in the wrong zone of their space. Most people are familiar with the phenomenon of the elevator: users look at the moving light panel because they do not wish to make eye contact with people who are standing closer than they would prefer. On buses, riders tense muscles that come into contact with others, don't make eye contact and do not speak when they might otherswise apologize in order to dehumanize (and aid others to dehumanize) other riders. Shutting out people in your bubble of personal space is a lot of effort and quite tiring.

Because the UBC bus loop will accomodate fluctuations in the number of people that will be using the plaza and adjacent spaces, it is inevitable that at some point people will be crowded closer to each other than they would prefer. In order to accomodate this, the space should be designed both sociofugally and sociopetally. In this way, people who wish to exclude others from their personal space could do so using less room than might otherwise be possible, but there would also be opportunities for groups to meet there comfortably. While sociofugal and sociopetal design is often seen as a matter of style, small moves can have an enormous effect.

People will chose adjacencies related to the different ways they are interacting. This can be broken down into three groups: people confront, consort, or coexist. When people are confronting each other, dominancy is unclear. These people will sit accross from each other to better judge facial expressions and body language. When people consort, it happens in two ways. When people collaborate by working on the same project, they will sit next to each other or around the corner of a table so they see the world from the same perspective. When people converse, they wish to see each other's facial expressions and so will face each other.

Most important is when people are coexisting. Most often people who are coexsting will seat themselves in a reciprocal arrangement that is diagonally separated to minimize eye contact. When trains or buses fill up, people who are coexisting will distribute themselves evenly throughout to minimize contact. In the case of a train, territories can be marked by a coat or briefcase.

This information is particularly useful to me because it begins to suggest ways in which MA's could be assembled to permit different kinds of interaction. The UBC bus loop should have space for confrontation, consorting and coexisting. Since there is not enough space for people to avoid entering each other's personal space, MA's should be arranged to allow people to coexist comfortably. One of the ways this can happen is by designing in sociopetal and sociopetal ways. This will allow spaces to remain ambiguous because the ways that people interact are not prescribed, but will be articulated by the combinations of MA's.

Thesis Advisor M remarked that while she liked the MA matrix, it was time to start thinking about making design decisions for specific reasons, instead of trying to find a program that would solve the problem itself (that's program in the computer sense, not the architectural sense). Arranging the MA's sociopetally and sociofugally would be a step towards doing this.

There are pros and cons to this. I'm arguing that the matrix increases ambiguity by combining surfaces while disregarding program and so creates infinite use opportunities rather than two or three use opportunities derived from the AGS. They are so weird it is not a choice of two or three, but a question of if it should be inhabited at all. If this is the case, I don't want to make assumptions about how people will be sitting because it starts to imply that I'm designing for the one or two programs. How can I mediate this problem?

I might just have to bite the bullet and say I'm designing for two or three uses.

Archigeek leaves for a short time for a consult

Team was out getting coffee, but Short Bus thinks that I should use both definitions of ambiguity. While I'm going to design for the uses specified in AGS, it's with the understanding that there are probably many other uses produced by the conjunction of surfaces that I can't control and don't want to. So, ambiguous is defined as 2+ or 3+ forms of occupation.

Also, there are some AGS's I combined that can't have sociopetal or sociofugal design applied to them. The envelope, ramp and stair do not allow inhabitation. In fact, I'm pretty sure the envelope shouldn't even be included, but it's too late now. The ramp and stair are about movement. You can sit on a stair, but you can't sit on the stair portion of a stair/seat, because the seat is defined as the part you sit on. As soon as you sit, the stair becomes a seat.

There are some AGS that already have relationships built in. The stage is designed for public space (near or far). The high work counter and the facing table allow for competing or consorting: while the distance is considered far-personal, I think the table/counter mitigates that relationship and changes it to close-social.

I think I'm going to take a look at the different relationships inherent in the AGS and see if that gets me anywhere.

Proxemics II

I've been reading up on proxemics: not just Edward Hall, but people who cite him, so that I can squeeze more use out what I've been looking at.

Before I start thinking, some definitions.

Proxemics: The study of how man structures microspace, how he relates physically to other persons with whom he is interacting and what is communicated by these physical relationships.

Fixed Feature: A permanent piece of a building. Always ceiling, walls and doors. In my project, seating, etc.

Semi-Fixed Features: Portions of a building that are always there, but that move around. Chairs, tables, etc. My project wants to make semi-fixed features completely fixed.

Informal Space: At least in this definition, the space that users leave stuff in to mark their territories.

Situational Behaviour: Behaviour that occurs when people enter 'personal space' bubbles that surround the body(distances appropriate to lovers, friends, and strangers are identified).

Territory: The area around themselves that a user wishes to remain free that is 'marked' can be done with a bag or jacket: size is related to relationships with adjacent people.

Sociofugal: A space that is designed to keep others away from the users personal space. Discourages interaction, but is gravitated towards by users who do not wish to interact.

Sociopetal: A space that is designed to encourage interaction with others. Allows groups to interact.

Mental Congruence: When a person feels that his or her personal characteristics, values and life stules are accommodated by particular spactial patterns.

Experiential congruence: How well the environment actually supports the functions, characteristics and behaviours of pepole using it.

Proxemics

Who says time spent socializing doesn't pay? I was talking with Team last night at the school Good Times about my project, and he suggested that one of the problems I may be having with the active axos is that I don't have enough people in them. I thought that was a very interesting suggestion, and came in early today to do a little research.

I came across a field called proxemics. Basically, proxemics studies how close people like to be to each other. So I've spent the day so far figuring out what the boundaries are and photoshopping/Cadding to explain some of my ideas. I'll post something as soon as it is coherent.

Prepping the TCM II

TCM being, of course, Thesis Committee Meeting.

I've been working on what I've been calling 'active axos'. I've taken the sectional moment drafted in AutoCAD and applied depth to it. Because the microarchitectural articulations (MA's) are still unrelated to program, the active axos illustrate the different depths associated with different levels of use.

So what I've done is chosen three (so far) MA's, and recreated them in Maya. They start with a thickness of 1", to show that they are derived from the sections. Right now, they are all the same colour, but it might be useful to show the different Graphic Standards sources in different colours. After 1", they change over 10 frames to the minimum spatial requirements for each Graphic Standard to be used by one person. In the next 10 frames they change to comfortable spatial requirements for one person. Frames 20-30 make the GS wide enough to be used by more than one person. Frames 40-50 show other: in some circumstances (like stairs and seats) it gets wider, the facing table gets access because it's limited to how far people will slide into the benches, etc. etc.

I've been thinking about the active axonometrics as the equivalent to a quadratic equation, and while I think they start out like that, they aren't showing as much information as I want them to. I'm not going to do that many, certainly: I thought 5 would be sufficient to show what I'm thinking. But while they start out great, there is something that just isn't working.

I need to sort this out before the end of the weekend because I need to put a presentation together for Tuesday. What's missing? What have I taken for granted that can be pulled apart to show something interesting?

UBC Statistics

Here are some things that may affect the spatial requirements of Microarchitectural Articulations:

Statistics Canada says that in Regional District Electoral Area A, 83.4 % of the populations is over 18. This means that I can stick to adult size articulations for design.

The report I got from Properties Trust says that at some times over 800 students are waiting for the B-Line. 4,500 people get off the bus between 9:00 and 10:00. The total UBC population in 2005 was 53,300.

This University Town brochure forecasts 10,700 University Town and Hampton Place residents in 2021, and 10,200 other campus residents in student residences and graduate housing. The projected future population is 20,900.

With this in mind, I think I'm going to design a space for certain ranges of people: 0-10, 10-25, 25-50, 50-100, 100-200, and 200-500 (I may not design for all of them, but instead pick a couple to investigate options)

It looks like the variations in size of people are negligeable, so I'm not going to worry about size fluctuations affecting use. The articulations are supposed to be ambiguous anyways, so while short people might have to find a different way to use them it should still be possible.

Research...

I've spent the last few days completing my matrix of Microarchitectural articulations (MA's). I'm now the architect of my own Matrix! Hooray!

I now have 115 iterations. The first generation is made up of Architectural Graphic Standards crossed with Architectural Graphic Standards and second generation is Architectural Graphic Standards crossed with the first iteration. I love the way the graph looks and I'm happy about finishing it.

The next step is applying some depth to these moments. Right now they are cool, but still very diagrammatic. The problem is that I don't have any guiding principles for what kind of depth should be used: I have a couple of plan views from Architectural Graphic Standards that show the right depth for a single person sitting, working at a desk and walking, and I also have information for more than one persone. The issue is that I have not yet assigned program to each of the moments. Since I don't know what each one of them is going to do I can't assign depth based on that. Instead, I'm going to put together a list of all the variable factors that could influence the depth of each section and create an equivalent of a quadratic equation for a couple of key articulations.

I have my toolbox of sections, and it's time to think about applying them to a site. While I'm not ready to look at the physical requirements of space, I think it's reasonable to think about site issues that affect design.

So, what are the variables that affect the three-dimensionalisation of these sectional moments?

1. Amount of people in the space: The UBC bus loop has busy times and quiet times. How many people are in this space at maximum occupancy? How do their physical needs vary from the needs of people at minimum occupancy?
2. Demographics of people using space: While the UBC bus loop is used primarily by students, there are some exceptions to this and I need to know what they are. Since the moments are based on human dimensions, variations in people should equate to variations in design.

There must be more than these two, but I'll start with them and see if anythin interesting comes up.

February!

The past couple of days have been short on posts because I've gone back to the CAD work. I've put together a matrix that allows me to track which Architectural Graphic Standards I've crossed and what the result is. I've also put together some images that show this lineage a little more graphically. Unfortunately, I can't get them to post properly.

I've devoted a fair amount of time to 'breeding' microarchitectural articulations (MA's) and now have 31 first generation (that is, two Architectural Graphic Standards (AGS) images combined) and 13 second generation (from a 1st generation MA and AGS combined) for a total of 44. Considering I started with 12, I'm pleased.

I'm creating these to get a 'kit of parts': the next step will be applying 'thickness' to the sections with the plan information from AGS. I would like to get this done before I speak to Mari on Tuesday, but I should be fine if I have enough work to show the general idea.